Hydrogen and oxygen stable isotope composition of water in metaschoepite mineralization on U3O8

Hydrogen and oxygen stable isotope composition of water in metaschoepite mineralization on U3O8

Journal Pre-proof Hydrogen and oxygen stable isotope composition of water in metaschoepite mineralization on U3O8 Erik J. Oerter, Michael Singleton, Z...

15MB Sizes 0 Downloads 22 Views

Journal Pre-proof Hydrogen and oxygen stable isotope composition of water in metaschoepite mineralization on U3O8 Erik J. Oerter, Michael Singleton, Zurong Dai, Amanda Deinhart, Melissa Thaw, M. Lee Davisson PII:

S0883-2927(19)30275-6

DOI:

https://doi.org/10.1016/j.apgeochem.2019.104469

Reference:

AG 104469

To appear in:

Applied Geochemistry

Received Date: 6 August 2019 Revised Date:

5 November 2019

Accepted Date: 7 November 2019

Please cite this article as: Oerter, E.J., Singleton, M., Dai, Z., Deinhart, A., Thaw, M., Davisson, M.L., Hydrogen and oxygen stable isotope composition of water in metaschoepite mineralization on U3O8, Applied Geochemistry (2019), doi: https://doi.org/10.1016/j.apgeochem.2019.104469. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Ltd.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46

Hydrogen and Oxygen Stable Isotope Composition of Water in Metaschoepite Mineralization on U3O8 Erik J. Oerter1*, Michael Singleton1, Zurong Dai1, Amanda Deinhart1, Melissa Thaw1,2, M. Lee Davisson1 1 2

Lawrence Livermore National Laboratory, 7000 East Avenue, Livermore, CA 94550, USA. University of California, Merced, 5200 Lake Road, Merced, CA 95340, USA.

* Corresponding Author email: [email protected]

Highlights We exposed U3O8 to air with constant relative humidity and known hydrogen and oxygen stable isotope composition of the water vapor for two periods of 90 days. We measured the stable isotopes of hydration water in metaschoepite that formed on the U3O8. δ18O values of the metaschoepite hydration water are likely to reflect that of the water vapor the sample was exposed to. It may be possible to discern information about the provenance and history of an oxidized U3O8 sample from oxygen stable isotope measurements of metaschoepite mineral hydration water. Key Words Mineral hydration water, humidity exposure experiments, nuclear forensics, stable isotope hydrology, isotope exchange Abstract When exposed to humidity in an oxidizing atmosphere hydrated uranium oxide grows as a secondary mineral on aged U3O8 and may incorporate the oxygen stable isotope signature of the water vapor into the secondary uranium oxide, as well as hydrogen and oxygen stable isotopes into any mineral hydration water. Because geospatial variation in δ2H and δ18O values of atmospheric humidity and precipitation is well understood, the H and O stable isotope composition of mineral hydration waters can give information on the environment of mineral formation. We exposed powdered U3O8 to humidity with known H and O stable isotope composition at constant 30%, 61% and 91% relative humidity. We sampled and analyzed the U3O8 powders along with any secondary hydrated minerals that had formed on the particle surfaces at various intervals from 1 to 10 days throughout the 180-day humidity exposures. We present stable H and O isotope results of mineral hydration waters in uranium oxide materials

Water Isotopes in Metaschoepite Mineralization on U3O8 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92

analyzed by thermogravimetry-enabled isotope ratio infrared spectroscopy (TGA-IRIS), which uses precise heating by thermogravimetric analyzer to liberate water vapor for subsequent online isotope analysis via a laser-based isotope ratio infrared spectroscopy instrument (Picarro L2130i). Scanning electron microscopy (SEM), transmission electron microscopy (TEM), and xray diffraction (XRD) analysis of humidity-exposed U3O8 heating to 350 °C indicates that the U3O8 had a measurable metaschoepite ((UO3)•2H2O) phase along with α-U3O8. After heating to 350 °C, the metaschoepite is eliminated but the α-U3O8 is retained, meaning that the heating successfully extracted the water in the metaschoepite but did not disturb the crystalline structure of the U3O8. Analysis of the bulk oxygen isotope composition by fluorination of these samples before and after heating, indicates that the oxygen isotope value of the U3O8 is also not affected by heating to 350 °C. Precision for δ2H and δ18O values of metaschoepite mineral hydration water stable isotope values yielded by the TGA-IRIS method on metaschoepite are 5.86‰ for δ2H, and 0.34‰ for δ18O. Hydrogen in the various water reservoirs is highly exchangeable and thus the δ2H values of these waters is not easily interpreted. However, oxygen in the separate water reservoirs is much less exchangeable, and thus δ18O values of the metaschoepite interlayer water is likely to reflect that of the water vapor the sample was exposed to. Once metaschoepite is formed on the surface of U3O8 particles, the oxygen isotope signature of the interlayer water in metaschoepite does not respond to changes in exposure vapor isotopic composition, and thus appears to be fairly durable in this regard. These results suggest that it may be possible to discern information about the provenance and history of an oxidized U3O8 sample from oxygen stable isotope measurements of metaschoepite mineral hydration water.

1. Introduction The need for the development of techniques to determine the provenance and history of nuclear materials has been identified (Moody et al., 2005; Mayer et al., 2012) and many recent efforts have focused on radioisotopic and elemental approaches (e.g. Eppich et al., 2013; Keegan et al., 2014; Horne et al., 2014). U3O8 is ubiquitous in the production of nuclear material, and one potentially attractive approach that has not yet been developed to determine the provenance and history of U3O8 is the analysis of the hydrogen and oxygen stable isotope composition of hydrated uranium oxidation weathering products that form from it. Hydrated mineral weathering products are common in general, and form from exposure to oxidative conditions in the environment due to an abundance of water and oxygen at ambient temperatures (e.g. Garrels and Christ, 1965; Maher and Chamberlain, 2014). Hydrated secondary minerals formed on U3O8 are also common (Tamasi et al., 2015; Lloyd et al., 2009; He et al., 2012) and many are included in the “schoepite group”, with general formulae of UO3•xH2O (Finch et al., 1992; Finch et al., 1996; Finch et al., 1998; Kubatko et al., 2006). Herein, we consider metaschoepite as UO3•2H2O (Tamasi et al., 2015). The mineral crystal structure of the schoepite group is orthorhombic with an arrangement of planar uranyl sheets interleaved with water molecules in the interlayer spaces (Finch et al., 1996; Plasil, 2018). The structural difference between schoepite and metaschoepite is much smaller than previously considered, leading to the easy transformation of schoepite to metaschoepite under ambient conditions and a joint occurrence of both minerals (Plasil, 2018). Because these hydrated minerals form in response to the oxygen and water they are exposed to, that water may be retained in the mineral crystal structure (Lawrence and Taylor, 1971; 1972). In this way they may develop and retain information on the environmental conditions of their formation because

2

Water Isotopes in Metaschoepite Mineralization on U3O8 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138

the parent water involved in the oxidation reactions may be of meteoric or ambient humidity origin, and therefore have geographically distinct hydrogen and oxygen stable isotope compositions (δ2H and δ18O values, defined below) (e.g. Dansgaard, 1964; Bowen, 2010). If the isotopic composition of this mineral hydration water can be measured, information about the mineral formation environment can be understood (e.g. Savin and Epstein, 1970; Shepard and Gilg, 1996; Savin and Hsieh, 1998). Until the recent advent of thermogravimetry-enabled isotope ratio infrared spectroscopy (TGA-IRIS; Oerter et al., 2017) the isotopic analysis of mineral hydration water was restricted by the labor-intensive off-line water extraction and gaseous conversion required by conventional magnetic sector mass spectrometry. In the case of isotope ratio infrared spectroscopy, isotope ratio measurements are made by the differential laser absorption of the heavy (ex. 2H) and light (1H) isotopologues of water vapor (with no gaseous conversion) that the laser pulse encounters. TGA-IRIS is an attractive analysis method because with minimal sample preparation, the sample can be step-heated very precisely to release waters held in the mineral matrix at different binding strengths (i.e. lower temperatures for weakly-bound, higher temperature for strongly-bound), and the corresponding mass loss of each water type can be precisely measured. Thus, TGA-IRIS allows for the rapid on-line analysis of large numbers of samples for simultaneous measurements of water content, δ2H and δ18O values. In the case of U3O8 secondary mineralization, as well as other minerals that contain multiple reservoirs of oxygen and/or hydrogen, TGA-IRIS offers the additional possibility to isotopically measure only the hydrous reservoirs and leave the oxygen in the mineral crystal structure unaltered. In this study, we exposed U3O8 to air with known water vapor content (relative humidity) and known hydrogen and oxygen stable isotope composition for two periods of 90 days. We periodically sampled the U3O8 through the exposure experiments and analyzed the secondary mineralization hydration water’s hydrogen and oxygen stable isotope composition by TGA-IRIS. The purposes of this study were to: 1) determine the feasibility and precision of TGA-IRIS measurements of mineral hydration water in secondary mineralization of U3O8, 2) determine the effects of differing levels of water vapor (relative humidity levels) on the mineral hydration water accumulation rates of U3O8 secondary mineralization, 3) determine the effects of differing levels of water vapor (relative humidity levels) on the δ2H and δ18O values of mineral hydration water in U3O8 secondary mineralization, and 4) determine the extent to which the isotopic composition of U3O8 secondary mineralization hydration waters are maintained after initial formation.

2. Experimental Methods 2.1 Uranium material A stock of powdered U3O8 (0.83 weight % 235U) was obtained from New Brunswick Laboratory (Chicago, IL, USA). Mineral composition of the U3O8 material before and after the humidity exposure experiments were determined by X-ray diffraction on a Bruker D8 Advance instrument (Bruker Corp., Billerica, MA, USA) using Cu-Kα radiation generated at 40 kV and 40 mA. Diffraction scans were performed from 10-80° 2θ, with 0.02° 2θ step size with 2 sec collection time per step, with variable divergence slits. 2.2 Humidity Exposure Experiments

3

Water Isotopes in Metaschoepite Mineralization on U3O8 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183

Three separate humidity exposure chambers were set up to create air with constant relative humidity (30%, 61%, 91%; ± 1.3%) and constant hydrogen and oxygen stable isotope composition of the water vapor (δ2Hvap, ± 1.2‰; and δ18Ovap, ± 0.15‰). Aqueous salt-saturated solutions to control humidity were not used to avoid the deleterious effects that salts have on the resulting water vapor (Oerter et al., 2018). The conceptual design of the humidity exposure chambers is based on mixing of wet and dry air streams, and more detail can be found in Oerter et al. (2019). Briefly, a stream of dry air is directed to a water reservoir where it passes through a water vapor-permeable membrane immersed in the water. This water vapor-saturated air stream is then joined by a diluter stream of dry air before the combined stream enters the humidity chamber. The relative proportion of vapor and diluter air streams are adjusted via mass flow controllers to achieve the desired humidity level inside each chamber, which are 10.5 L in volume. Temperature and relative humidity (RH) inside the chambers are monitored by high precision digital thermometer and hygrometer (± 0.2 °C, ± <1% RH; HMT-330, Vaisala Inc., Finland), and relative humidity (RH) data are reported here. Each chamber received a total air flow of 500 standard cubic centimeters per minute (SCCM), corresponding to 3 air changes per hour, an air flow rate that is designed to emulate that of a typical warehouse ventilation system (ANSI, 2013). Two isotopically-distinct waters were used to supply water vapor: TWW with δ2Hliq = -111.32 ‰ and δ18Oliq = -14.98 ‰ (average δ2Hvap = -186.3 ‰ and δ18Ovap = -24.3 ‰ at average temperature of 23.8 °C); and KD with δ2Hliq = -0.92 ‰ and δ18Oliq = -0.01 ‰ (average δ2Hvap = -80.4 ‰ and δ18Ovap = -8.9 ‰ at average temperature of 22.9 °C). Hydrogen and oxygen stable isotope values are reported in δ notation: δ = (Rsample/Rstandard – 1), where Rsample and Rstandard are the 2H/1H or 18O/16O ratios for the sample and standards respectively, and values are reported in per mille (parts per thousand, ‰). Water vapor δ values are calibrated to the Vienna Standard Mean Ocean Water (VSMOW) scale (Coplen, 1994) by analyzing liquid water standards at the beginning and end of each sample set. Aliquots of ~3 g of U3O8 powder were contained in open 20 mL glass vials placed inside each chamber. The U3O8 powder was distributed evenly across the bottom of each vial, forming a disc ~2 cm in diameter, with ~1 mm thickness, though this volume diminished throughout the exposure experiments as the powders were sampled. Two separate humidity exposure experiments were conducted: 1) exposure of U3O8 to KD-vapor for 90 days, and 2) remaining material from the 90-day KD vapor experiments were transferred to TWW-vapor chambers for another 90 days of TWW-vapor exposure. 2.3 SEM and TEM analytical methods Aliquots of U3O8 from all humidity exposure levels were analyzed after the exposure experiments by scanning electron microscopy (SEM) to reveal the morphology and texture of the powders. Aliquots of material were dispersed on carbon adhesive tabs mounted on standard SEM pin stubs (Ø12.7mm x 8mm pin height), and SEM was performed with a JEOL JSM-7401F FESEM at an accelerating voltage of 5 kV. Transmission electron microscopy (TEM) was performed after the exposure experiments on the material from the 91% RH exposure to reveal microstructural detail of the powders. TEM thin section specimens were prepared by focused ion beam (FIB) technique using a FEI Nova 600 dual beam microscope comprising a Ga+ liquid metal source FIB and a field emission gun SEM. A Pt strap was applied onto the top surface of a TEM slice by in-situ deposition induced firstly by electron beam (e-Pt coating) and then by ion beam (i-Pt coating) to protect the material

4

Water Isotopes in Metaschoepite Mineralization on U3O8 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229

surface structure from bombardment damage during FIB cutting. TEM analysis was done using a FEI 80-300kV Titan transmission electron microscope operating at 300 kV.

2.4 TGA-IRIS analytical methods To make δ2H and δ18O measurements of secondary mineralization hydration water, a TA Discovery thermogravimetric analyzer (TGA) (TA Instruments, New Castle, DE, USA) with infrared-heated furnace and 25 position sample changer was connected to a Picarro L-2130i cavity ring down isotope ratio infrared spectroscopy (IRIS) water isotope analyzer (Picarro Inc., Santa Clara, CA, USA) by a sample transfer line heated to 80 °C (Oerter et al., 2017). Water vapor generated by sample heating in the TGA is carried through the system by N2 carrier gas, flowing at 30 mL min-1. Duplicate samples of U3O8 were taken from each humidity chamber at Day 0, 1, 2, 5, 8, 11, 15, 20, 29, 40, 50, 59, 70, 78, 90, and single samples at Day 91, 92, 94, 97, 100, 105, 111, 120, 146, 160, and 180. At each sampling event, aliquots of U3O8 (60 to 80 mg) were loaded into pre-tared aluminum sample holders and immediately hermetically crimp-sealed with lids to prevent exposure to ambient humidity (<5 min exposure time). All TGA-IRIS analyses began with the autosampler opening each sample holder lid, loading the sample into the TGA furnace, and immediately closing the furnace. The TGA furnace was held at 35 °C and purged with dry N2 for 5 minutes to flush the system of any ambient water vapor and return [H2O] values in the IRIS instrument to ≤ 200 ppmV. To first identify the U3O8 material’s water release temperature ranges for subsequent isotope measurements, an initial aliquot was heated at a constant rate of 10 °C min-1 from 35 to 500 °C (“ramp heated”). Based on the ramp heating results (discussed more below), samples for hydrogen and oxygen stable isotope analysis were heated at a very fast rate of ≥ 25 °C sec-1 (“step heated”) to rapidly release all of the water available in each temperature range as a single pulse of sufficient peak size for reliable IRIS oxygen and hydrogen stable isotope analysis. The isothermal dwell time at each heating step was 10 minutes for complete dehydration at each temperature. The same samples were then step heated to 350 °C to release mineral hydration water, again followed by 10 min isothermal dwell time. 2.4 Fluorination and IRMS analytical methods To make measurements of the total oxygen isotope composition in a sample (structural oxygen in U3O8 and UO3•2H2O, and hydration water in UO3•2H2O), which we denote as δ18OTotal, we analyzed samples of U3O8 with varying amounts of metaschoepite (UO3•2H2O) present (specific samples described more below) by fluorination and isotope ratio mass spectrometry (IRMS). Samples of 30 to 40 mg where placed into nickel crucibles inside nickel reaction vessels and pre-fluorinated with 140 torr ClF3 at ~25 °C for 5 minutes. This pre-fluorination step was conducted to remove any adsorbed water vapor on the sample or any reaction products inside the nickel vessels resulting from sample loading. The prefluorination products are pumped away and 590 torr ClF3 was loaded into each nickel reaction vessel and heated at 550 °C for 4 hours. Liberated O2 was passed through two LN2 traps to remove any condensable gases and a heated NaCl trap to remove F2, and then collected on 13X molecular sieve at LN2 temperature. The 13X sieve was then warmed to room temperature, and the O2 was transferred to an IsoPrime-100 spectrometer operating in dual-inlet mode. No standard reference materials are currently available for oxygen 5

Water Isotopes in Metaschoepite Mineralization on U3O8 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254

isotope compositions in U3O8. Values of δ18OTotal were calibrated relative to VSMOW based on analysis of the silica sand standard NBS-28 (quartz, δ18OTotal = 9.6‰) which were analyzed in duplicate during each fluorination run. Precision for δ18OTotal values determined on standard deviation of the calibration standards was ±0.11‰. 3. Results 3.1 Thermal water release profiles To assess the H2O release profiles during TGA heating for uranium oxide samples, and to determine the appropriate step-heating scheme for subsequent water vapor isotope analysis, we heated an aliquot of the U3O8 starting material in the TGA (without isotope analysis) from 35 to 500 °C at a constant rate (ramp heating) of 10 °C min-1. Ramp heating at 10 °C min-1 of this material resulted in a two-step mass loss profile, with initial mass loss occurring from ~50 to ~105 °C, followed by another interval from ~250 to 350 °C (Fig. 1). After the humidity exposures, the samples had similar mass release profiles. Based on these mass loss profiles with ramp heating, we determined the heating scheme to release the water available at each temperature step in a single pulse for isotope analyses to be step-heating first to 105 °C followed by 10 minutes isothermal at 105 °C, and then step-heating to 350 °C followed by 10 minutes isothermal at 350 °C.

100.00

500

A

450

99.75

400 99.50 350

Weight (%)

U3O8 sample #1253

300

99.00

250 200

(°C)

B

98.75

Temperature

99.25

150 98.50 100 98.25

50

98.00

0 0

255 256 257 258 259 260

10

20

30 Time

40

50

60

(min)

Figure 1. Mass loss (% relative to starting sample mass) for a sample of the starting U3O8 material (blue line) with constant heating from 35 to 500 °C at 10 °C min-1 (green line); (A) represents mass loss from ~50 to 105°C, and (B) represents mass loss from ~250 to 350 °C.

6

Water Isotopes in Metaschoepite Mineralization on U3O8 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278

3.2 Water accumulation with humidity exposure The accumulation of water on the sample particle surfaces from exposure to humid air can be measured as an increase in mass loss during TGA heating. Mineral hydration rates were determined as water mass accumulation rates calculated from linear regressions of the data arrays of mass loss upon heating shown in Figure 2, assuming all mass difference from the start of the exposure experiments was accumulated water. Water accumulation rates for the water yielded at 105 °C and 350 °C are shown in Table 1. U3O8 in our experiments only accumulated water at 91% RH (Fig. 2). The rates of water mass accumulation for U3O8 available at 350 °C are approx. 50% to 60% of that available at 105 °C (Table 1). Table 1. Water mass accumulation rates for U3O8 samples in the humidity exposure experiments expressed as % mass loss day-1 upon heating in TGA-IRIS analysis. Only those with non-zero water accumulation rates are shown.

279 280 281

1.40 1.30 1.20 1.10 1.00 0.90 0.80 0.70 0.60 0.50

30% RH 61% RH 91% RH

0

282 283 284 285 286 287 288 289 290 291 292

B 1.50

U3O8 - 105° C

20 40 60 80 100 120 140 160 180 200

Days of Exposure

Mass Loss (%)

Mass Loss (%)

A 1.50

U3O8 - 350° C 30% RH 61% RH 91% RH

1.40 1.30 1.20 1.10 1.00 0.90 0.80 0.70 0.60 0.50 0

20 40 60 80 100 120 140 160 180 200

Days of Exposure

Figure 2. Plots of the mass loss upon heating in the TGA-IRIS to (A) 105 °C and (B) 350 °C for U3O8 exposed to KD water vapor for 90days, and then TWW water for another 90 days. 3.3 Secondary mineralization from humidity exposure Analysis by XRD of the U3O8 material before humidity exposure show that the starting material was dominated by α-U3O8, with minor metaschoepite (UO3⋅2H2O) present. This minor pre-existing secondary mineralization is likely from intermittent ambient humidity exposure since material synthesis in the 1980’s. After 90 days of exposure, the samples of U3O8 from all of the humidity levels were re-examined by XRD and all had measurable metaschoepite present.

7

Water Isotopes in Metaschoepite Mineralization on U3O8 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337

After the second 90 day exposure (180 days total) all samples had metaschoepite still present. There was no evidence of transformation of metaschoepite to schoepite. A comparison of the SEM analyses of the samples exposed to 30%, 61%, and 91% RH for 180 days shows that platy-tabular crystals of secondary mineralization occur on the surfaces of the U3O8 particles (Fig. 3). Due to the lack of crystal structural information from the SEM observations, we are not able to unambiguously determine from the SEM observations that these crystals are strictly the result of secondary mineralization from humidity exposure. In addition, the U3O8 material itself has some layered structural morphology, further increasing the difficulty in resolving the secondary mineralization by SEM imaging. However, the abundance and size of these tabular mineral crystals does correlate with increased exposure humidity (Fig. 3). To further determine the nature of the secondary mineralization formed during humidity exposure, we examined a sample of starting material (unexposed U3O8) and the U3O8 sample exposed to 91% RH for 180 days total (first KD-water vapor for 90 days then TWW-water vapor for another 90 days) by using TEM imaging and electron diffraction analysis. Tabular shaped secondary mineral phases are identified from both samples in addition to the primary micrometer-sized U3O8 crystals (Fig. 4A and 4F). In the sample of starting material (unexposed U3O8), the secondary phase predominantly formed on the surfaces of U3O8 particles and occasionally bridged U3O8 particles (Figs. 4A-4C). Characteristic selected-area electron diffraction patterns taken from the larger, micrometer-sized particles match that of single α-U3O8 crystals with good crystallinity (Figs. 4D and 4I). In the humidity-exposed sample, more abundant additional secondary phases grew between the U3O8 particles, displaying distinctive tabular shapes that are sub-micrometer in thickness (Figs. 4F-4H). Bright-field (Fig. 4G) and dark-field (Fig. 4H) TEM images show a tabular secondary mineralization crystal that grew between U3O8 particles. Growth of these tabular secondary mineral phases are a signature of uranium oxide that has been subjected to high-humidity exposure (Donald et al., 2017). The electron diffraction pattern taken from the secondary phase formed on the surface of the starting material (unexposed U3O8) displays a polycrystalline ring-pattern overlapping a diffuse scattering of an amorphous matrix (Fig. 4E), of which the polycrystalline ring-pattern matches that of UO2, inferring alteration products of schoepite or metaschoepite. In the case of the humidity-exposed U3O8 sample, an electron diffraction pattern from a sub-micrometer sized tabular secondary mineralization phase displays a more predominant ring pattern diffraction feature in which the primary reflection rings match polycrystalline UO2 (Fig. 4J). Importantly, extra reflections matching that of metaschoepite are also present near to the central transmitted spot (inset of Fig. 4J). These extra reflections have d-spacings of 0.73, 0.52, 0.40 and 0.36 nm, corresponding to increasing distance from the central transmitted spot, which can be attributed to a mixture of schoepite/metaschoepite (Finch et al., 1996; Finch et al., 1997; Finch et al., 1998) and dehydrated schoepite, likely as γ-uranyl dihydroxide (Siegel et al., 1972). We also analyzed by XRD a sample of the U3O8 that had been exposed to KD-water vapor at 91% RH for 90 days, both before and after TGA-IRIS heating and analysis at 350 °C. Before heating, the U3O8 sample that had been exposed to 91% RH for 90 days had a metaschoepite component along with the primary α-U3O8 (Fig. 5). After heating to 350 °C for TGA-IRIS analysis, the metaschoepite was eliminated but the α-U3O8 remained and displayed no crystal structure transformation (Fig. 5).

8

Water Isotopes in Metaschoepite Mineralization on U3O8 338 339 340 341

342 343 344 345 346 347 348 349 350 351 352

Figure 3. Secondary electron SEM images showing typical morphology of U3O8 powders: (A) starting material, and material exposed to (B) 30%, (C) 61%, and (D) 91% RH KD-water vapor for 90 days, then switched to TWW-water vapor for another 90 days (180 days total humidity exposure).

9

Water Isotopes in Metaschoepite Mineralization on U3O8

353 354 355 356 357 358 359 360 361 362 363 364 365

Figure 4. TEM observation and electron diffraction analysis of the sample before exposure (AE) and the sample exposed to 91% RH for 180 days (F-J). (A) and (F) are bright-field TEM images of FIB sections where arrowheads indicate the alteration products of mateschoepite or schoepite formed either on the surface of or between U3O8 particles. (B) and (C) are bright-field TEM images of a closer looking of the alteration product of metschoepite formed on the surface of U3O8. (D) and (E) are selected-area electron diffraction patterns collected from U3O8 (circled in white on (B)) and alteration products of schoepite (circled in white on (C)). Bright-field (G) and dark-field (H) TEM images show a closer looking of tabular schoepite/metaschoepite crystal that grew between U3O8 particles. (I) and (J) are selected-area electron diffraction patterns collected from the U3O8 (circled in white at lower-left corner on (H)) and the tabular product (circled in white at upper-right corner on (H)). The ring patterns (E) and (J) are indexed

10

Water Isotopes in Metaschoepite Mineralization on U3O8 366 367 368 369 370 371

372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391

matching crystal structure of UO2. Besides the reflection rings of UO2, some extra reflections can be identified in (J), as indicated by the thicker arrowheads on the inset which is an enlarged pattern of the part enclosed in the white dash-line box.

Figure 5. Phase analysis by XRD from pre- and post-heating of 91% RH-exposed U3O8 to 350°C during TGA-IRIS analysis. Red line shows XRD diffraction pattern of U3O8 material before heating, black arrows indicate diagnostic metaschoepite reflections, and black line shows XRD diffraction pattern of the same sample post-350 °C heating without metaschoepite and no transformation of U3O8 crystal structure.

3.4 Stable isotope measurements by TGA-IRIS 3.4.1 Precision of isotopic measurements We estimate the internal precision, or the variability between replicate measurements of the same material, by calculating the standard deviation of the TGA-IRIS isotope measurements of the two replicate samples of each material taken during each sampling event, and these results are shown in Table 2. This approach captures the variability between TGA-IRIS measurements made on each material that was exposed to differing humidity levels, with differing exposure duration, as well as the variability in measurements made on water yielded at 105 °C and 350 °C. For U3O8 measurements made at 105 °C the internal precision for δ2H ranged from 3.33‰ to 6.01‰, and for δ18O was 0.69‰ to 0.84‰ (Table 2). For measurements made at 350 °C δ2H values ranged from 3.55‰ to 5.86‰, and for δ18O was 0.20‰ to 0.33‰. No clear trends 11

Water Isotopes in Metaschoepite Mineralization on U3O8 392 393 394 395 396 397 398 399 400 401 402 403 404

405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431

in the internal precision figures with humidity levels are present for U3O8. No standard reference materials for the isotopic composition of mineral hydration waters are available. However, the external precision on goethite synthesized with water of known isotopic composition using the same instrument and analytical approach used here were reported as 0.3‰ to 1.6‰ for δ2H, and 0.17‰ to 0.27‰ for δ18O (Oerter et al., 2017). Based on these various metrics, we take the most conservative values and estimate the overall precision of TGA-IRIS measurements made on metaschoepite on U3O8 at 105 °C as 6.01‰ for δ2H values, 0.84‰ for δ18O values, and at 350 °C as 5.92‰ for δ2H, and 0.33 for δ18O. Table 2. Estimated internal precision of TGA-IRIS isotope measurements of metaschoepite, calculated as the average of standard deviations calculated from pairs of measurements through the first 90 day humidity exposures.

3.4.2 Stable isotopic composition of metaschoepite hydration water The δ2H and δ18O values of water recovered from U3O8 exposed to KD-vapor for 90 days are shown in Figure 6. Water yielded by TGA heating at 105 °C and 350 °C have δ2H values that exhibited broadly similar trends, though with different δ2H values, through the humidity exposures: 1) relatively low δ2H values before the material was placed in the humidity chambers (Day 0), 2) then an increase of approximately 50‰ to maxima near Day 20, 3) followed by a slight decrease through the end of the exposures (Figs. 6A and 6B). There was not a clear influence of the δ2H composition of the exposure water vapor on the δ2H values of the metaschoepite hydration water. There were also no clear systematic differences in δ2H values of metaschoepite hydration water between the various humidity exposures, except that δ2H values from the 91% RH exposures tended to be higher than for 61% or 30% RH. When the exposure water vapor for U3O8 was switched from high δ2H values (KD-vapor) to lower δ values (TWW-vapor), the δ2H values of the water metaschoepite hydration water yielded at both 105 °C and 350 °C showed distinct step-like decreases towards the isotopic values of the new exposure water vapor (Figs. 6A and 6B). The step decreases were of a similar magnitude in δ2H values for both 105 °C and 350 °C but were not as large as the decrease in water vapor δ2H values and did not match that of the new water vapor. For δ18O values of metaschoepite hydration water yielded at 105 °C from U3O8 exposed first to KD-vapor for 90 days, a similar pattern to the δ2H values of the same material was found: low at the start, higher near Day 20, and a slow decrease through the end of the exposure, with no clear correspondence with δ18O values of the exposure water vapor (Fig. 6C). In contrast, δ18O values of metaschoepite hydration water yielded at 350 °C exhibited a distinct trend through the exposure: δ18O values of the starting material (Day 0) were ca. -13‰ and metaschoepite hydration water yielded by TGA heating of material from all humidity exposures

12

Water Isotopes in Metaschoepite Mineralization on U3O8 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446

showed increasing δ18O values up to Day 15 (Fig. 6D), after which the isotopic values either levelled out and showed clear correspondence with the water vapor, in the case of 91% RH, or had isotopic values slightly lower than the water vapor for the 61% and 30% RH exposed samples. The 350 °C δ18O values also had lower variability between replicates measured on the same day, compared to either 105°C δ18O values or any δ2H values. When the exposure water vapor for the U3O8 was switched from high δ18O values (KDvapor) to lower δ values (TWW-vapor), the δ18O values of the metaschoepite hydration water yielded at 105 °C showed a distinct step-like decrease towards the isotopic values of the new exposure water vapor but still did not match that of the new water vapor (Fig. 6C). For metaschoepite hydration water yielded at 350 °C δ18O values showed a slight decrease but did not approach that of the new water vapor for the rest of the 90 day exposure (Fig. 6D). There was a systematic difference between different exposure humidities, with δ18O values of water from the 91% RH humidity showing less shift after the vapor change and remaining at higher δ18O values than water from the 61% or 30% exposure humidities.

13

Water Isotopes in Metaschoepite Mineralization on U3O8 447

448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468

Figure 6. Isotopic results of metaschoepite hydration water formed on U3O8 exposed to KDvapor for 90 days, and then exposed to TWW-vapor for an additional 90 days. Values of δ2H and δ18O of water from samples of U3O8 heated to 105 °C (A, C) and 350 °C (B, D) analyzed by TGA-IRIS. Starting material average δ2H and δ18O values are shown as white bars at Day 0. Water vapor data shown is from 61% RH exposure chamber. Precision of isotopic values are shown in Table 2.

3.5 O isotopic composition by fluorination and IRMS To evaluate the O isotope effects of TGA heating on the underlying U3O8 and coexisting metaschoepite, we analyzed the δ18O values of all of the oxygen (interlayer water and uranylbound, δ18OTotal) in the unexposed U3O8 starting material (Day 0) and of U3O8 samples exposed to KD-vapor for 90 days, followed by exposure to TWW-vapor for another 90 days (Day 180), by fluorination and IRMS. Importantly, δ18OTotal does not include any water adsorbed to particle surfaces because the samples were pre-flourinated to remove any water in the fluorination system. Aliquots of each sample type were analyzed before and after heating to 350 °C during TGA-IRIS analysis (n=2 replicates for each sample type). The results are shown in Table 3. The samples that represent the unexposed U3O8 starting material (Day 0) had average δ18OTotal values of 12.0‰, and there was not a difference (within analytical precision) between

14

Water Isotopes in Metaschoepite Mineralization on U3O8 469 470 471 472 473 474 475 476 477

478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505

the unheated material samples, and those that had been heated to 350 °C. For the samples that had been exposed to 91% RH for 180 days, the average δ18OTotal value of the unheated samples was 10.89‰, and was 9.85‰ for samples that had been heated to 350 °C (Table 3).

Table 3. Values of δ18OTotal from fluorination and IRMS analyses.

4. Discussion 4.1 Secondary mineralization of U3O8 under humidity exposure The secondary phases observed by TEM and electron diffraction formed on and between the primary U3O8 particles are secondary mineralization products composed of metaschoepite. These products were formed by U3O8 reacting with H2O through sorption, oxidation, dissolution and precipitation processes under an oxidizing environment in the humidity exposure experiments. During TGA-IRIS analysis, mass lost in the 50 to 105 °C interval represents dehydration of weakly-adsorbed water on mineral surfaces (Foldvari, 2011; Oerter et al., 2017). Although there is no increase in mass loss, and thus a lack of surface-sorbed water accumulation at 30% and 61% RH, compared to the linear increase in surface-sorbed water with exposure time for the samples exposed to 91% RH (Fig. 2A), the isotope values of these desorbed waters show a similar pattern (Figs. 6A and 6C). These similar isotope values suggest that this water originates from the same source, or combination of sources. Under a low or middle humidity, it will take a much longer time to form metaschoepite because of a limitation in the amount of H2O adsorbed on the surface of U3O8 and slow dissolution rate of the U3O8. The case observed for the starting material belongs to this case, in which the minor metaschoepite on the surface of the starting material U3O8 formed during a long period of material storage. The inference in the case of the unexposed U3O8 starting material is case is that the U3O8 particles have not been subjected to a high humidity environment. The increase in the amount of surface-sorbed water for the 91% RH samples (Fig. 2) suggests that this extra water is stored on the particle surfaces in a different form than for the lower humidity samples. Indeed, humidity levels of ~30% correspond to 1 monolayer of water 15

Water Isotopes in Metaschoepite Mineralization on U3O8 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551

adsorbed to particles surfaces, ~60% to 2 monolayers, and ~90% to 5 monolayers. This variable amount of water present at the particle surfaces may lead to the formation of surface hydroxyls to different degrees and at different rates, as has been shown for various uranium oxide minerals (Finch et al., 1992). At the 91% RH level, there are likely to be localized aqueous-saturated environments present on the particle surfaces, which result in the nucleation and growth of hydrated uranium minerals (Finch et al., 1992; Donald et al., 2017). In the case observed for the sample exposed to 91% RH for 180 days, the hydrated phase was metaschoepite, which formed by reacting with H2O supplied as vapor in the exposure chambers. These various surface hydration and mineral precipitation mechanisms, along with the water accumulation trends shown in Fig. 2 suggest that there is a humidity threshold between 61% and 91% RH that leads to increased surface water accumulation and hydrated mineral formation on U3O8 particles. Analyzing a sample by XRD that had been exposed to 91% RH both before and after heating to 350 °C allowed the mineralogical source of the water yielded at that temperature during TGA-IRIS analysis to be determined. Before heating, the sample had a distinct metaschoepite component along with α-U3O8 (Fig. 5). After heating to 350 °C, the metaschoepite was eliminated but the α-U3O8 remained and was not affected by the heating (Fig. 4). Thus, mass lost during heating in the 250 to 350 °C interval (Fig 1) is water held within the metaschoepite crystal structure in the interlayers between sheets of uranyl ions. In the case of the UO2 crystallites associated with the metaschoepite in the starting material (Fig. 4E), these may have formed by phase transformation from the metaschoepite during material storage due to possible variations of humidity and temperature. In the 180-day exposure at 91% RH sample the UO2 crystallites associated with the secondary tabular metaschoepite phases (Fig. 4J) may have formed by phase transformation from the metaschoepite that was induced under focused ion beam bombardment during TEM sample preparation. The metaschoepite is also sensitive to electron beam illumination. We are, therefore, unable to exclude that some of the UO2 crystallites could be an artefact of sample preparation or by electron beam illumination during TEM analysis. Although the initial crystal structure of metaschoepite is unable to be completely retained in TEM analysis, its morphology is retained and distinguishable from that of the primary U3O8 phase in TEM imaging, and it is therefore useful for investigation of the secondary mineralization. 4.2 Evaluation of isotope exchange between mineral water reservoirs A central question in the use of TGA-IRIS to analyze hydrogen and oxygen stable isotope signatures of hydration water in uranium oxide materials is: To what degree do the uranium oxides phases undergo chemical and physical changes during TGA-IRIS analysis, and do these physicochemical changes result in hydrogen and/or oxygen isotope exchange? The lack of crystal structure alteration of the U3O8 after heating indicates that neither the surface-sorbed water yielded upon heating at 105 °C, nor the metaschoepite structural water yielded at 350 °C interacts with or alters the U3O8 crystal structure of the material itself during TGA-IRIS analysis. In terms of the stable isotope composition of metaschoepite hydration water, the simultaneous presence of more than one “reservoir” of hydrogen and oxygen on and in the U3O8metaschoepite mineral complexes, indicates that it is important to evaluate the degree to which each element can move between each reservoir. If exchange can occur between each reservoir, and if the reservoirs have different isotopic compositions, with respect to each element, then the measured isotopic composition will not be indicative of the source of the water within the mineral.

16

Water Isotopes in Metaschoepite Mineralization on U3O8 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597

In metaschoepite hydrogen exists as water adsorbed to particle surfaces and as interlayer water between the uranyl sheets (Plasil, 2018). Oxygen in metaschoepite exists in these reservoirs, but also as oxygen bonded to uranium in the uranyl sheets. Isotopic exchange could occur between these reservoirs for each element during or after mineral formation, as well as during heating at 105 °C and then 350 °C for TGA-IRIS analysis. A simple test for isotope exchange is comparison between values of the water vapor yielded at each temperature, provided that the isotopic composition is different in each reservoir. If exchange has occurred there will be a correlation between δ values of the water vapor yielded at each temperature step. For hydrogen, there are strong correlations between δ2H values yielded at 105 °C and at 350 °C that ranges from R2 = 0.94 to 0.99, with slopes from 0.92 to 1.05, and intercepts from 20.5 to -10.0 (Fig. 7A). The tight correlations and slopes that are parallel to each other and the 1:1 relationship between the two temperatures indicate that there is hydrogen isotope exchange between water adsorbed on particle surfaces and that held in interlayers during heating for TGAIRIS analysis. However, it appears from linear data arrays that are offset from 1:1 that there is a difference in the isotopic composition of each water reservoir within the mineral, and that there is not exchange during or after mineral formation. The fractionation factor for adsorbed water appears to be 10‰ to 20‰ higher than that of the water held in interlayer spaces (Fig. 7A). Mixing of the two reservoirs during TGA-IRIS heating does not explain the apparent fractionation because the amount of water in the reservoirs would have to be the same to result in such consistent offsets from 1:1, and the water yields increase through time for the 91% RH exposures (Fig. 2). For oxygen, there are not correlations between δ18O values yielded at 105 °C and at 350 °C, with R2 = 0.33 to 0.58 (Fig. 7B), indicating that there is not oxygen isotope exchange between water adsorbed on particle surfaces and that held in interlayers during heating for TGAIRIS analysis. However, the data arrays of the KD-vapor exposed samples do plot about 6‰ above the 1:1 line in Figure 7B, though with considerable scatter, indicating that the oxygen fractionation factor of the interlayer water is about 6‰ higher than that of the adsorbed water. The samples that were switched into the TWW-vapor after KD-vapor exposure have a larger deviation from the 1:1 line, but this is probably due to the adsorbed water (105 °C) showing more response to the change to TWW-vapor with low δ18O values (discussed more below). 4.3 OTotal isotopic composition by fluorination and IRMS The total O isotope composition (structural oxygen in U3O8 and UO3•2H2O, and hydration water in UO3•2H2O), of the Day 0 samples, was 12.0‰ and was the same for unheated Day 0 samples, and those that had been heated to 350 °C. This similarity in δ18OTotal values indicates that: (A) heating to 350 °C does not have an effect on the O isotope composition of the underlying U3O8 material, and (B) that the pre-existing metaschoepite on the starting material was of such small quantity that its OTotal isotopic composition did not have a measurable influence on that of the whole sample U3O8 δ18OTotal value. For the samples that had been exposed to water vapor at 91% RH and had newly-formed metaschoepite on the particle surfaces, the average δ18OTotal value was 10.89‰, which is 1.13‰ lower than the Day 0 samples. The lower δ18OTotal value of the Day 180 samples probably results from the inclusion of water from the exposure humidity with lower δ18O values into the hydration water of the metaschoepite (UO3•2H2O). There could also be oxygen with lower δ18O values from the water vapor included into the structural oxygen of the UO3 group in the metaschoepite, which was re-incorporated into the U3O8 of the Day 180 samples after 17

Water Isotopes in Metaschoepite Mineralization on U3O8 598 599 600 601 602 603 604 605 606 607

608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628

dehydration at 350 °C. This water vapor-derived oxygen in the heated Day 180 samples leads to an average δ18OTotal value of 9.85‰, which is 1.04‰ lower than that of the non-heated Day 180 samples, and 2.19‰ lower than the Day 0 samples. This incorporation of water vapor-derived oxygen into the residual U3O8 material from the 91% RH-exposed samples, but not the Day 0 samples may be a matter of there not being enough water available in the minor pre-existing metaschoepite on the Day 0 samples to contribute enough oxygen to be detectable. Indeed, the 91% RH exposed samples do not begin to accumulate significant metaschoepite hydration water until about Day 20 of the humidity exposures (Figure 2B).

Figure 7. Comparison of δ values of water released from U3O8 at 105 °C and 350 °C for (A) δ2H and (B) δ18O from 90 day water vapor exposure experiments with KD-vapor (“KD”), and 180 day isotope reversal experiment from KD- to TWW-vapor (“KD → TWW”). Solid line is 1:1. 4.4 Hydration water isotopic composition and its durability The exchangeability of hydrogen between adsorbed and interlayer water, and the lack thereof for oxygen helps explain some of the patterns observed in the δ values of the humidity exposure experiments. In this discussion, we assume that the water released at 105 °C represents the adsorbed water and that the interlayer water is released at 350 °C. The initial rise towards higher δ2H values for the first 20 days in all of the humidity exposures (Fig. 6C) could be explained by mixing of the adsorbed water with some water with higher δ2H values (and then exchange with the interlayer water during TGA-IRIS analysis). However, there is not a water source identified in this study with such a high δ2H value. Some insight is gained from the gradual return to the KD-vapor in the exposures from 20 to 90 days, as was expected. We suggest that this early rise-gradual return pattern could be explained by a kinetic isotope fractionation effect happening at the particle surfaces. If the surfaces are not water saturated, as would be expected at the beginning of the humidity exposures, there could be a kinetic isotopic fractionation similar to evaporation taking place as water at the particle surface is not in

18

Water Isotopes in Metaschoepite Mineralization on U3O8 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674

equilibrium with the surrounding vapor. This effect would diminish through time as the two water reservoirs come into equilibrium and the effect is minimized. When the switch to TWW-vapor occurred at 90 days, there was a distinct step towards 2 lower δ H values of both adsorbed and interlayer water, though not to the full extent of the change in exposure vapor (Figs. 7C and 7D). The lesser shift could be explained by preexisting water not fully being replaced by the new vapor. In the case of hydrogen, the remarkably similar patterns through time of the adsorbed (105 °C) and interlayer (350 °C) water δ2H values are explained by the exchangeability of hydrogen during TGA-IRIS heating. For oxygen, the pattern in the adsorbed δ18O values is very similar to that of δ2H, suggesting similar processes are occurring, which supports the kinetic fractionation effect due to the disequilibrium between adsorbed water and exposure vapor scenario discussed above. After the switch from KD-vapor to TWW-vapor at 90 days, the adsorbed water δ18O values almost approach that of the new exposure vapor, but don’t ever fully match, supporting the incomplete replacement of preexisting adsorbed water. The δ18O values of interlayer (350 °C) water display the most organized response to exposure humidity and attain isotopic equilibrium with the exposure humidity in about 15 days for 91% RH, which is maintained for the rest of the KD-vapor exposure (Fig. 6B). The lower humidity exposures show a similar shift towards the exposure vapor δ18O values within 20-25 days, but never fully match, and may move to slightly lower δ18O values through the 90 day exposures, though variability in the data make it hard to resolve a distinct shift (Fig. 6B). After the switch to TWW-vapor with lower δ values at 90 days, the interlayer water δ18O values show a subtle shift lower, but of only a few ‰ and maintain these new values throughout the remainder of the exposures. The difference between the δ18O values from the 91% RH exposures, and that of the 61% and 30% RH exposures (which are essentially the same), suggests that there is a humidity threshold between 61% and 91% in the response of the oxygen stable isotope composition of the interlayer water in metaschoepite to the relative humidity it forms in. Below 61% the oxygen stable isotope composition of metaschoepite does not fully attain that of the exposure atmosphere, but at 91% RH it does, suggesting that higher humidity is required to form new metaschoepite. The constant δ18O values of interlayer water in metaschoepite throughout the humidity exposures regardless of humidity level suggest that once formed, metaschoepite δ18O values are stable and durable, even if the oxygen isotope composition of the exposure humidity changes. 5. Conclusions We determined that when U3O8 is exposed to humid air, metaschoepite secondary mineralization will form on the surface of the U3O8 particles at high humidity levels. The hydrogen and oxygen stable isotope composition of the water adsorbed to the particle surfaces and that incorporated into the mineral as hydration water is measurable by TGA-IRIS without affecting the underlying U3O8 material. Hydrogen in the various water reservoirs is highly exchangeable and thus the δ2H data of these waters is not easily interpreted. However, oxygen in the separate water reservoirs is much less exchangeable, and thus δ18O values of the metaschoepite interlayer water is likely to reflect that of the water the sample was exposed to. The oxygen isotope signature of the interlayer water in metaschoepite also appears to be fairly durable, and once formed does not respond to changes in exposure vapor isotopic composition. These results suggest that it may be possible to discern information about the provenance and history of an oxidized U3O8 sample from oxygen stable isotope measurements of

19

Water Isotopes in Metaschoepite Mineralization on U3O8 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720

metaschoepite mineral hydration water. Further work to discern the effects of exposure temperature on the oxygen stable isotope composition of metaschoepite hydration water and the durability of the signature to changes in both temperature and isotopic composition should be undertaken. Of particular interest is the relation between the oxygen isotope compositions of vapor and the incipient formation of metaschoepite on stoichiometric U3O8. Data Availability https://doi.org/10.7910/DVN/W3UKPW Acknowledgements We thank R. Thaw for providing the TWW water. This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344, and was funded by the Office of Defense Nuclear Nonproliferation Research and Development within the U.S. Department of Energy's National Nuclear Security Administration under Project Numbers LL15-U_Surface_Oxidation-NDD3B and LL18-U Surface Oxidation Part II-NDD3Bb. Release number LLNL-JRNL-773024.

References Bowen, G.J. (2010) Isoscapes: spatial pattern in isotopic biogeochemistry. Annual Review of Earth and Planetary Sciences 38, 161-187. Coplen, T.B. (1994) Reporting of stable hydrogen, carbon, and oxygen isotopic abundances (technical report). Pure and Applied Chemistry 66, 273-276. Dansgaard, W. (1964) Stable Isotopes in Precipitation. Tellus 16, 436-468. Donald, S.B., Dai, Z.R., Davisson, M.L., Jeffries, J.R. and Nelson, A.J. (2017) An XPS study on the impact of relative humidity on the aging of UO2 powders. Journal of Nuclear Materials 487, 105-112. Eppich, G.R., Williams, R.W., Gaffney, A.M. and Schorzman, K.C. (2013) 235 U–231 Pa age dating of uranium materials for nuclear forensic investigations. Journal of Analytical Atomic Spectrometry 28, 666-674. Finch, R.J., Cooper, M.A., Hawthorne, F.C. and Ewing, R.C. (1996) The crystal structure of schoepite,[(UO 2) 8 O 2 (OH) 12](H 2 O) 12. The Canadian Mineralogist 34, 1071-1088.

Finch, R. J., Hawthorne, F. C., Miller, M. L., and Ewing, R. C. (1997). Distinguishing among schoepite,[(UO 2) 8 O 2 (OH) 12](H 2 O) 12, and related minerals by X-ray powder diffraction. Powder Diffraction, 12(4), 230-238. Finch, R.J., Hawthorne, F.C. and Ewing, R.C. (1998) Structural relations among schoepite, metaschoepite and" dehydrated schoepite". The Canadian Mineralogist 36, 831-845.

20

Water Isotopes in Metaschoepite Mineralization on U3O8 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766

Finch, R.J., Miller, M.L. and Ewing, R.C. (1992) Weathering of natural uranyl oxide hydrates: schoepite polytypes and dehydration effects. Radiochimica Acta 58, 433-444. Földvári, M. (2011) Handbook of thermogravimetric system of minerals and its use in geological practice. Geological Institute of Hungary Budapest. Garrels, R.M. and Christ, C.L. (1965) Solutions, minerals, and equilibria. He, H., Wang, P., Allred, D.D., Majewski, J., Wilkerson, M.P. and Rector, K.D. (2012) Characterization of chemical speciation in ultrathin uranium oxide layered films. Analytical chemistry 84, 10380-10387. Horne, S., Landsberger, S. and Dickson, B. (2014) Determination of isotopic ratios of uranium samples using passive gamma spectroscopy with multiple detectors. Journal of Radioanalytical and Nuclear Chemistry 299, 1171-1175. Keegan, E., Kristo, M.J., Colella, M., Robel, M., Williams, R., Lindvall, R., Eppich, G., Roberts, S., Borg, L. and Gaffney, A. (2014) Nuclear forensic analysis of an unknown uranium ore concentrate sample seized in a criminal investigation in Australia. Forensic science international 240, 111-121. Kubatko, K.-A., Helean, K., Navrotsky, A. and Burns, P.C. (2006) Thermodynamics of uranyl minerals: Enthalpies of formation of uranyl oxide hydrates. American Mineralogist 91, 658-666. Lawrence, J. and Taylor, H. (1972) Hydrogen and oxygen isotope systematics in weathering profiles. Geochimica et Cosmochimica Acta 36, 1377-1393. Lawrence, J.R. and Taylor, H.P. (1971) Deuterium and oxygen-18 correlation: Clay minerals and hydroxides in Quaternary soils compared to meteoric waters. Geochimica et Cosmochimica Acta 35, 993-1003. Lloyd, N.S., Mosselmans, J.F.W., Parrish, R.R., Chenery, S.R.N., Hainsworth, S.V. and Kemp, S.J. (2009) The morphologies and compositions of depleted uranium particles from an environmental case-study. De Gruyter. Maher, K. and Chamberlain, C. (2014) Hydrologic regulation of chemical weathering and the geologic carbon cycle. Science 343, 1502-1504. Mayer, K. (2013) Security: Expand nuclear forensics. Nature News 503, 461. Mayer, K., Wallenius, M. and Varga, Z. (2012) Nuclear forensic science: correlating measurable material parameters to the history of nuclear material. Chemical reviews 113, 884-900. Moody, K.J., Grant, P.M., Hutcheon, I.D. and Varoufakis, Y. (2014) Nuclear forensic analysis. CRC Press.

21

Water Isotopes in Metaschoepite Mineralization on U3O8 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799

Oerter, E., Singleton, M. and Davisson, L. (2017) Hydrogen and oxygen stable isotope signatures of goethite hydration waters by thermogravimetry-enabled laser spectroscopy. Chemical Geology 475, 14-23. Oerter, E.J., Singleton, M. and Davisson, M.L. (2018) Hydrogen and oxygen stable isotope dynamics of hyper-saline and salt-saturated aqueous solutions. Geochimica et Cosmochimica Acta. Oerter, E.J., Singleton, M., Thaw, M. and Davisson, M.L. (2019) Water vapor exposure chamber for constant humidity and hydrogen and oxygen stable isotope composition. Rapid Communications in Mass Spectrometry 33, 89-96. Plasil, J. (2018) The crystal structure of uranyl-oxide mineral schoepite, [(UO2)4O(OH)6](H2O)6, revisited, Journal of Geosciences, 63, 65-73. Savin, S.M. and Epstein, S. (1970) The oxygen and hydrogen isotope geochemistry of clay minerals. Geochimica et Cosmochimica Acta 34, 25-42. Savin, S.M. and Hsieh, J.C. (1998) The hydrogen and oxygen isotope geochemistry of pedogenic clay minerals: principles and theoretical background. Geoderma 82, 227-253. Sheppard, S. and Gilg, H. (1996) Stable isotope geochemistry of clay minerals. Clay Minerals 31, 1-24. Siegel, S., H. R. Hoekstra, E. Gebert (1972) The Structure of -Uranyl Dihydroxide, UO2(OH)2, Acta Crystallographica, B28, 3469-3473. Tamasi, A.L., Boland, K.S., Czerwinski, K., Ellis, J.K., Kozimor, S.A., Martin, R.L., Pugmire, A.L., Reilly, D., Scott, B.L. and Sutton, A.D. (2015) Oxidation and hydration of U3O8 materials following controlled exposure to temperature and humidity. Analytical chemistry 87, 4210-4217.

22

Material U3O8

Experiment

RH

% loss day-1 @ 105 °C

R2

% loss day-1 @ 350 °C

R2

KD 90 day

91%

0.0028

0.95

0.0015

0.77

U3O8

KD -> TWW 180 day

91%

0.0026

0.94

0.0016

0.90

105 °C 350 °C δ2H St. Dev. ‰ δ18O St. Dev. ‰ δ2H St. Dev. ‰ δ18O St. Dev. ‰

Material U3O8

Experiment

RH

KD 90 day

30%

3.33

0.69

3.55

0.33

U3O8

KD 90 day

61%

6.01

0.84

5.92

0.23

U3O8

KD 90 day

91%

5.80

0.81

5.86

0.20

Unheated

After TGA-IRIS heating to 350 °C

Replicate

δ OTotal ‰

δ18OTotal ‰

U3O8, Starting Material

1

12.06

11.96

U3O8, Starting Material

2 Average

11.98 12.02

12.13 12.04

U3O8, KD -> TWW, Day 180

1

10.95

9.98

U3O8, KD -> TWW, Day 180

2 Average

10.83 10.89

9.73 9.85

Sample

18

Declaration of interests ☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. ☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: